The purpose and objectives of this study were to estimate the current and future health and economic impacts of major mental illnesses on Canadian society over the next 30 years. To achieve these objectives RiskAnalytica’s Life at Risk simulation platform was used to generate a base model of the current and future life and economic outcomes associated with major mental illness in Canada including: mood disorders, anxiety disorders, schizophrenia, disorders of childhood and adolescence, cognitive impairment including dementia, and substance use disorders.
Using data inputs from a wide range of sources, the base model estimates the current impact of major mental illnesses and forecasts it over a 30-year time horizon. This base model represents the status quo for each mental illness type, where measures of health (incidence, prevalence and mortality) and economics (direct and indirect costs attributable to each illness) can be computed. It is important to note that the model assumes that the baseline prevalence and incidence of mental illness will remain constant over time it is further assumed that health service use and productivity will remain constant over the simulated time period. As a result, the projections are driven primarily by changes in population demographics. Based on current knowledge, this is considered to be a reasonable assumption for the majority of mental illnesses considered in this study, however the estimates presented are considered conservative if the incidence of one or more disorders increases or liberal if decreases in incidence are observed.
The base model takes into account some key risk factors and protective factors such as the healthy immigrant effect, comorbidity among mental illnesses, comorbidity between mental illnesses and two chronic diseases (type II diabetes and heart disease) and the increased risk of mental illness in adulthood given a childhood or adolescent illness. Each is described in further detail in Section 2.
Out of Scope Components of and Limitations to the Current Model
This study only considered the life and economic burden for the major illnesses listed in Section 1.1. The impacts of other mental illnesses (such as eating disorders, personality disorders, and other additions such as gambling) were not included as a result of data limitations and time constraints. However these disorders are less common within the population and tend to co-occur with the major mental illnesses included in this study. As a result the impact of omitting these illnesses on the overall findings is considered to be minimal.
In addition, each mental illness was modeled as a distinct diagnostic category and the severity of illness and subclinical illnesses were not taking into consideration within this analysis. In reality the symptoms of each illness can occur across a continuum from mild to severe. At the mild end of the spectrum, diagnostic criteria may not be met, but mild cases may be at risk of future illness that meets diagnostic criteria, especially among children and youth (Galanter and Patel, 2005). In other situations, despite meeting diagnostic criteria, individuals may cope well and the impact of their illness in terms of suffering, disability, health care service use, and workplace productivity may be minimal (WHO, 2003). It
is important to note that the costs and health care service utilization rates presented within this report were not adjusted for severity of illness and the probability of health service use based on illness severity.
Similarly, there are many known risk factors for mental illnesses that have not been included such as socioeconomic status, adverse childhood events and genetic factors. Due to data and time constraints, the model only included age, sex, type II diabetes, heart disease, immigrant status, and previous or concurrent mental illness as risk factors. Finally, it was not possible to include a full range of outcomes for some disorder. In particular, for children and youth disorders, outcomes such as leaving school early, teen pregnancy and parenting, domestic violence, injury, vehicle collision, fatality rates and legal system involvement (Eggeret al. 2003; Roweet al. 2004) were not considered.
The total costs presented in this report could not be specifically and directly attributed to a particular mental illness type or illness severity. In addition, the current model does not include costs for the justice system, education system, the provision of informal care, and the use of excess health services for conditions other than those provided for the main condition. These additional costs may comprise a substantial proportion of the total economic burden of mental illness. The utilization rate analysis for direct costs associated with mental illness was limited to the utilization and costs data available in Canada which is a subset of the total direct costs and mental illnesses included in the study.
The indirect costs associated with mental illness were limited to adult mental illness types and did not consider the lost opportunity costs for informal caregivers for time off work to care for the elderly, children and adolescents, or those with severe mental illness.
Finally, this study presents the impact of mental illness in Canada from a national perspective and does not take into consideration differences in specific population groups such as First Nations, Inuit and Métis (FNIM) or specific jurisdictions such as provinces and territories, in terms of epidemiological risk, health equity and access, and health services and resources available.
While this study considered many data sources, major challenges to producing an optimal model were substantial limitations in the type, age, availability, relevancy and quality of datasets in Canada. The data gaps and limitations are outlined in detail, and assumptions used to address data limitations were discussed and approved by the SME panel and considered to provide a reasonable approximation based
on current knowledge. While acknowledging data limitations and uncertainties, the Life at Risk model has allowed, for the first time, connections among data sets for a reasonable and comprehensive picture of the impact of mental illness in Canada. Please refer to Sections 2.8 and 4.2 for further discussion on data and study limitations respectively.
1.3 OVERVIEW OF MENTAL ILLNESSES